Mobile Vision API 

Tato Kutalia tatocaster tatocaster.me github.com/tatocaster twitter.com/@tatokutalia 

What is Vision API? 

What is Vision API? Find objects in photos and video, using real-time on-device vision technology. • detect faces • scan barcodes • recognize text 

Face Detection • not recognition, just detection • The Face API finds human faces in photos, videos, or live streams. It also finds and tracks positions of facial landmarks such as the eyes, nose, and mouth • classification 

Face Orientation 

Landmarks Landmark detection is not done by default, since it takes additional time to run. You can optionally specify that landmark detection should be done. 

Classification Classification determines whether a certain facial characteristic is present. The Android Face API currently supports two classifications: eyes open and smiling. but iOS Face API currently supports only smiling classification. note: “eyes open” and “smiling” classification only works for frontal faces, that is, faces with at most about +/- 18 degrees. 

Face tracking extends face detection to video sequences. Any face appearing in a video for any length of time can be tracked. That is, faces that are detected in consecutive video frames can be identified as being the same person. Note: This is not face recognition 

1. Create 

2. Detect 

3. Release 

Result 

Operational Status The first time that an app using the Face API is installed on a device, GMS will download a native library to the device in order to do face detection. A detector’s isOperational method can be used to check if the required native library is currently available 

Some weird things • detect nose base • scale bitmap • calculate Y for external bitmap • action 

No content

Barcode • 1D barcodes: EAN-13, EAN-8, UPC-A, UPC-E, Code-39, Code-93, Code-128, ITF, Codabar • 2D barcodes: QR Code, Data Matrix, PDF-417, AZTEC Android 4.2.2 

Text Recognition • The Text API can recognize text in any Latin based language Real-Time, on device 

Text Recognition The Text Recognizer segments text into blocks, lines, and words. a Block is a contiguous set of text lines, such as a paragraph or column, a Line is a contiguous set of words on the same vertical axis, and a Word is a contiguous set of alphanumeric characters on the same vertical axis. 

No content

more • mobile OCR codelab : 
 https://codelabs.developers.google.com/codelabs/mobile-vision-ocr/ • GitHub samples 
 https://github.com/googlesamples/android-vision 

more • https://github.com/tatocaster/VisionAPIAndRxJava • https://speakerdeck.com/tatocaster/mobile-vision-api 

No content